Kacheong Poon writes:
> Henk Langeveld wrote:
> 
> > I have a concern about checksum off-load - how is e2e integrity guarded?
> > This may have been beaten to death already, but I just wonder how it is
> > handled.
> 
> 
> If an app is really concerned about e2e integrity, it should
> do its own integrity checking.  The TCP checksum is pretty
> weak after all.

Or use a transport such as SCTP that's stronger.

One of the trade-offs of checksum off-load is that you can't guard
against errors that occur somewhere between the network interface
device itself (where the generation/check takes place) and the
transport layer.

It's always been the case that there's an unprotected path from the
transport layer up through the application itself.  So, at least in a
strict definition, we're not talking about end-to-end anyway, but
rather degrees of protection.  (No application just 'ends' at TCP.)

If you were hoping that TCP checksums would protect against (say) DMA
errors between the network device and the stack itself, that
protection is disabled by off-load.  The trade-off you get is not
having to waste CPU time looking at every byte against the
unlikelihood of such (system design) errors.

-- 
James Carlson, KISS Network                    <[EMAIL PROTECTED]>
Sun Microsystems / 1 Network Drive         71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to